Automated AI research poses an unacceptable risk of losing control to superhuman systems

Artificial intelligence research, products, and impact across society.

The field of preventing harm from advanced AI systems, spanning alignment, evaluation, and decisions to slow or withhold models whose capabilities outpace their safeguards.

The process by which an AI system improves its own capabilities, enabling increasingly effective further improvements.

The hypothesized rapid, AI-driven acceleration of AI progress in which automated AI research compresses years of advances into months or less.

The scientific study and development of artificial intelligence systems, algorithms, and methodologies.